This Jupyter notebook project uses YOLOv8 for vehicle tracking and implements a line crossing detection algorithm. The system counts vehicles that cross a specified line in a video, annotates the frames, and generates an output video with visualizations.
- OpenCV
- Ultralytics YOLO
- supervision
- Clone the repository:
git clone https://github.com/shaadclt/Vehicle-Counter-YOLOv8.git cd Vehicle-Counter-YOLOv8
- Open the Jupyter notebook:
jupyter notebook
- Run the 'vehicle_counter.ipynb' notebook.
- Adjust line coordinates: 'START' and 'END' in the notebook.
- Configure YOLOv8 model parameters as needed.
The output video output_single_line.mp4 will be generated with annotated frames showing object tracks and the count of objects that crossed the line.
- This project uses the YOLOv8 model from Ultralytics.